arXiv subjects

Yan Liu

Publications and source records attributed to Yan Liu.

At least 433 records · Page 24Linked to original sources

arXiv2018

Can I trust you more? Model-Agnostic Hierarchical Explanations

Interactions such as double negation in sentences and scene interactions in images are common forms of complex dependencies captured by state-of-the-art machine learning models. We propose Mahé, a novel approach to provide Model-agnostic hierarchical éxplanations of how powerful machine learning models, such as deep neural networks, capture these interactions as either dependent on or free of the context of data instances. Specifically, Mahé provides context-dependent explanations by a novel local interpretation algorithm that effectively captures any-order interactions, and obtains context-free explanations through generalizing context-dependent interactions to explain global behaviors. Experimental results show that Mahé obtains improved local interaction interpretations over state-of-the-art methods and successfully explains interactions that are context-free.

stat.ML↗

arXiv2018

Trace class perturbation of closed linear relations

This paper studies trace class perturbation of closed linear relations in Hilbert spaces. The concept of trace class perturbation of closed relations is introduced by orthogonal projections. Equivalent characterizations of compact and trace class block operator matrices are first given in terms of their elements, separately. By using them, several equivalent and sufficient characterizations of trace class perturbation of closed linear relations are obtained.

math.SP↗

arXiv2018

Single SiV$^-$ centers in low-strain nanodiamonds with bulk-like spectral properties and nano-manipulation capabilities

We report on the isolation of single SiV$^-$ centers in nanodiamonds. We observe the fine-structure of single SiV$^-$ center with improved inhomogeneous ensemble linewidth below the excited state splitting, stable optical transitions, good polarization contrast and excellent spectral stability under resonant excitation. Based on our experimental results we elaborate an analytical strain model where we extract the ratio between strain coefficients of excited and ground states as well the intrinsic zero-strain spin-orbit splittings. The observed strain values are as low as best values in low-strain bulk diamond. We achieve our results by means of H-plasma treatment of the diamond surface and in combination with resonant and off-resonant excitation. Our work paves the way for indistinguishable, single photon emission. Furthermore, we demonstrate controlled nano-manipulation via atomic force microscope cantilever of 1D- and 2D-alignments with a so-far unreached accuracy of about 10nm, as well as new tools including dipole rotation and cluster decomposition. Combined, our results show the potential to utilize SiV$^-$ centers in nanodiamonds for the controlled interfacing via optical coupling of individually well-isolated atoms for bottom-up assemblies of complex quantum systems.

cond-mat.mes-hall↗

arXiv2018

Dial2Desc: End-to-end Dialogue Description Generation

We first propose a new task named Dialogue Description (Dial2Desc). Unlike other existing dialogue summarization tasks such as meeting summarization, we do not maintain the natural flow of a conversation but describe an object or an action of what people are talking about. The Dial2Desc system takes a dialogue text as input, then outputs a concise description of the object or the action involved in this conversation. After reading this short description, one can quickly extract the main topic of a conversation and build a clear picture in his mind, without reading or listening to the whole conversation. Based on the existing dialogue dataset, we build a new dataset, which has more than one hundred thousand dialogue-description pairs. As a step forward, we demonstrate that one can get more accurate and descriptive results using a new neural attentive model that exploits the interaction between utterances from different speakers, compared with other baselines.

cs.CL↗

arXiv2018

Skeleton-based Activity Recognition with Local Order Preserving Match of Linear Patches

Human activity recognition has drawn considerable attention recently in the field of computer vision due to the development of commodity depth cameras, by which the human activity is represented as a sequence of 3D skeleton postures. Assuming human body 3D joint locations of an activity lie on a manifold, the problem of recognizing human activity is formulated as the computation of activity manifold-manifold distance (AMMD). In this paper, we first design an efficient division method to decompose a manifold into ordered continuous maximal linear patches (CMLPs) that denote meaningful action snippets of the action sequence. Then the CMLP is represented by its position (average value of points) and the first principal component, which specify the major posture and main evolving direction of an action snippet, respectively. Finally, we compute the distance between CMLPs by taking both the posture and direction into consideration. Based on these preparations, an intuitive distance measure that preserves the local order of action snippets is proposed to compute AMMD. The performance on two benchmark datasets demonstrates the effectiveness of the proposed approach.

cs.CV↗

arXiv2018

An application of $L^m-L^r$ estimates to weakly coupled systems of semilinear viscoelastic wave equations

We consider weakly coupled systems of semilinear viscoelastic wave equations with different power source nonlinearities in $\mathbb{R}^n$, $n\geq1$ as follows: \begin{equation*} \left\{\begin{aligned} &u_{tt}-Δu+g\astΔu+u_t=|\partial_t^{\ell}v|^p,\\ &v_{tt}-Δv+g\astΔv+v_t=|\partial_t^{\ell}u|^q,\\ \end{aligned}\right. \end{equation*} with $\ell=0,1$ and $p,q>1$. After presenting $L^m-L^r$ estimates with $1\leq m\leq r\leq \infty$ of solutions to the corresponding linearized problem with vanishing right-hand side, we prove the existence of global in time solutions to the weakly coupled systems, where the initial data are supposed to belong to different $L^r$ spaces with different additional $L^m$ regularities.

math.AP↗

arXiv2018

Weyl semimetal/insulator transition from holography

We study a holographic model which exhibits a quantum phase transition from the strongly interacting Weyl semimetal phase to an insulating phase. In the holographic insulating phase there is a hard gap in the real part of frequency dependent diagonal conductivities. However, the anomalous Hall conductivity is nonzero at zero frequency, indicting that it is a Chern insulator. This holographic quantum phase transition is always of first order, signified by a discontinuous anomalous Hall conductivity at the phase transition, in contrast to the very continuous holographic Weyl semimetal/trivial semimetal phase transition. Our work reveals the novel phase structure of strongly interacting Weyl semimetal.

hep-th↗

arXiv2018

Building a Controlled Vocabulary for Standardizing Precision Medicine Terms

Rapid advances of technology and development of research in precision medicine domain have led to the production of different types of biomedical data. Standard medical vocabularies were shown to be limited in dealing with such heterogeneous data and consequently, new controlled vocabulary for data integration and normalization has been proposed. In this study, the precision medicine vocabulary (PMV), which is a controlled vocabulary for terms used in precision medicine, is built based on the method of data integration in Unified Medical Language System (UMLS). It now covers ten top semantic types of disease, drug, gene, gene variation and so on. In total of 1,372,967 concepts and 4,567,208 terms have been integrated from widely used databases related with precision medicine.

cs.CY↗

arXiv2018

DynGEM: Deep Embedding Method for Dynamic Graphs

Embedding large graphs in low dimensional spaces has recently attracted significant interest due to its wide applications such as graph visualization, link prediction and node classification. Existing methods focus on computing the embedding for static graphs. However, many graphs in practical applications are dynamic and evolve constantly over time. Naively applying existing embedding algorithms to each snapshot of dynamic graphs independently usually leads to unsatisfactory performance in terms of stability, flexibility and efficiency. In this work, we present an efficient algorithm DynGEM based on recent advances in deep autoencoders for graph embeddings, to address this problem. The major advantages of DynGEM include: (1) the embedding is stable over time, (2) it can handle growing dynamic graphs, and (3) it has better running time than using static embedding methods on each snapshot of a dynamic graph. We test DynGEM on a variety of tasks including graph visualization, graph reconstruction, link prediction and anomaly detection (on both synthetic and real datasets). Experimental results demonstrate the superior stability and scalability of our approach.

cs.SI↗

arXiv2018

Feature Selection of Post-Graduation Income of College Students in the United States

This study investigated the most important attributes of the 6-year post-graduation income of college graduates who used financial aid during their time at college in the United States. The latest data released by the United States Department of Education was used. Specifically, 1,429 cohorts of graduates from three years (2001, 2003, and 2005) were included in the data analysis. Three attribute selection methods, including filter methods, forward selection, and Genetic Algorithm, were applied to the attribute selection from 30 relevant attributes. Five groups of machine learning algorithms were applied to the dataset for classification using the best selected attribute subsets. Based on our findings, we discuss the role of neighborhood professional degree attainment, parental income, SAT scores, and family college education in post-graduation incomes and the implications for social stratification.

cs.CY↗

arXiv2018

Deep Generative Dual Memory Network for Continual Learning

Despite advances in deep learning, neural networks can only learn multiple tasks when trained on them jointly. When tasks arrive sequentially, they lose performance on previously learnt tasks. This phenomenon called catastrophic forgetting is a fundamental challenge to overcome before neural networks can learn continually from incoming data. In this work, we derive inspiration from human memory to develop an architecture capable of learning continuously from sequentially incoming tasks, while averting catastrophic forgetting. Specifically, our contributions are: (i) a dual memory architecture emulating the complementary learning systems (hippocampus and the neocortex) in the human brain, (ii) memory consolidation via generative replay of past experiences, (iii) demonstrating advantages of generative replay and dual memories via experiments, and (iv) improved performance retention on challenging tasks even for low capacity models. Our architecture displays many characteristics of the mammalian memory and provides insights on the connection between sleep and learning.

cs.LG↗

arXiv2018

Anomalous transport model with axial magnetic fields

The transport properties of massless fermions in $3+1$ spacetime dimension have been in the focus of recent theoretical and experimental research. New transport properties appear as consequences of chiral anomalies. The most prominent is the generation of a current in a magnetic field, the so-called chiral magnetic effect leading to an enhancement of the electric conductivity (negative magnetoresistivity). We study the analogous effect for axial magnetic fields that couple with opposite signs to fermions of different chirality. We emphasize local charge conservation and study the induced magneto-conductivities proportional to an electric field and a gradient in temperature. We find that the magnetoconductivity is enhanced whereas the magneto-thermoelectric conductivity is diminished. As a side result we interpret an anomalous contribution to the entropy current as a generalized thermal Hall effect.

cond-mat.mes-hall↗

arXiv2018

Detecting Statistical Interactions from Neural Network Weights

Interpreting neural networks is a crucial and challenging task in machine learning. In this paper, we develop a novel framework for detecting statistical interactions captured by a feedforward multilayer neural network by directly interpreting its learned weights. Depending on the desired interactions, our method can achieve significantly better or similar interaction detection performance compared to the state-of-the-art without searching an exponential solution space of possible interactions. We obtain this accuracy and efficiency by observing that interactions between input features are created by the non-additive effect of nonlinear activation functions, and that interacting paths are encoded in weight matrices. We demonstrate the performance of our method and the importance of discovered interactions via experimental results on both synthetic datasets and real-world application datasets.

stat.ML↗

arXiv2018

Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting

Spatiotemporal forecasting has various applications in neuroscience, climate and transportation domain. Traffic forecasting is one canonical example of such learning task. The task is challenging due to (1) complex spatial dependency on road networks, (2) non-linear temporal dynamics with changing road conditions and (3) inherent difficulty of long-term forecasting. To address these challenges, we propose to model the traffic flow as a diffusion process on a directed graph and introduce Diffusion Convolutional Recurrent Neural Network (DCRNN), a deep learning framework for traffic forecasting that incorporates both spatial and temporal dependency in the traffic flow. Specifically, DCRNN captures the spatial dependency using bidirectional random walks on the graph, and the temporal dependency using the encoder-decoder architecture with scheduled sampling. We evaluate the framework on two real-world large scale road network traffic datasets and observe consistent improvement of 12% - 15% over state-of-the-art baselines.

cs.LG↗

arXiv2018

A Modified Sigma-Pi-Sigma Neural Network with Adaptive Choice of Multinomials

Sigma-Pi-Sigma neural networks (SPSNNs) as a kind of high-order neural networks can provide more powerful mapping capability than the traditional feedforward neural networks (Sigma-Sigma neural networks). In the existing literature, in order to reduce the number of the Pi nodes in the Pi layer, a special multinomial P_s is used in SPSNNs. Each monomial in P_s is linear with respect to each particular variable sigma_i when the other variables are taken as constants. Therefore, the monomials like sigma_i^n or sigma_i^n sigma_j with n>1 are not included. This choice may be somehow intuitive, but is not necessarily the best. We propose in this paper a modified Sigma-Pi-Sigma neural network (MSPSNN) with an adaptive approach to find a better multinomial for a given problem. To elaborate, we start from a complete multinomial with a given order. Then we employ a regularization technique in the learning process for the given problem to reduce the number of monomials used in the multinomial, and end up with a new SPSNN involving the same number of monomials (= the number of nodes in the Pi-layer) as in P_s. Numerical experiments on some benchmark problems show that our MSPSNN behaves better than the traditional SPSNN with P_s.

cs.LG↗

arXiv2018

Robustness of classification ability of spiking neural networks

It is well-known that the robustness of artificial neural networks (ANNs) is important for their wide ranges of applications. In this paper, we focus on the robustness of the classification ability of a spiking neural network which receives perturbed inputs. Actually, the perturbation is allowed to be arbitrary styles. However, Gaussian perturbation and other regular ones have been rarely investigated. For classification problems, the closer to the desired point, the more perturbed points there are in the input space. In addition, the perturbation may be periodic. Based on these facts, we only consider sinusoidal and Gaussian perturbations in this paper. With the SpikeProp algorithm, we perform extensive experiments on the classical XOR problem and other three benchmark datasets. The numerical results show that there is not significant reduction in the classification ability of the network if the input signals are subject to sinusoidal and Gaussian perturbations.

stat.ML↗

arXiv2017

Two-dimensional solitons and quantum droplets supported by competing self- and cross-interactions in spin-orbit-coupled condensates

We study two-dimensional (2D) matter-wave solitons in spinor Bose-Einstein condensates (BECs) under the action of the spin-orbit coupling (SOC) and opposite signs of the self- and cross-interactions. Stable 2D two-component solitons of the mixed-mode (MM) type are found if the cross-interaction between the components is attractive, while the self-interaction is repulsive in each component. Stable solitons of the semi-vortex type are formed in the opposite case, under the action of competing self-attraction and cross-repulsion. The solitons exist with the total norm taking values below a collapse threshold. Further, in the case of the repulsive self-interaction and inter-component attraction, stable 2D self-trapped modes, which may be considered as quantum droplets (QDs), are created if the beyond-mean-field Lee-Huang-Yang (LHY) terms are added to the self-repulsion in the underlying system of coupled Gross-Pitaevskii equations. Stable QDs of the MM type, of a large size with an anisotropic density profile, exist with arbitrarily large values of the norm, as the LHY terms eliminate the collapse. The effect of the SOC term on characteristics of the QDs is systematically studied. We also address the existence and stability of QDs in the case of SOC with mixed Rashba and Dresselhaus terms, which makes the density profile of the QD more isotropic. Thus, QDs in the spin-orbit-coupled binary BEC are for the first time studied in the present work.

cond-mat.quant-gas↗

arXiv2017

Kerr-Sen Black Hole as Accelerator for Spinning Particles

It has been proved that arbitrarily high-energy collision between two particles can occur near the horizon of an extremal Kerr black hole as long as the energy $E$ and angular momentum $L$ of one particle satisfies a critical relation, which is called the BSW mechanism. Previous researchers mainly concentrate on geodesic motion of particles. In this paper, we will take spinning particle which won't move along a timelike geodesic into our consideration, hence, another parameter $s$ describing the particle's spin angular momentum was introduced. By employing the Mathisson-Papapetrou-Dixon equation describing the movement of spinning particle, we will explore whether a Kerr-Sen black hole which is slightly different from Kerr black hole can be used to accelerate a spinning particle to arbitrarily high energy. We found that when one of the two colliding particles satisfies a critical relation between the energy $E$ and the total angular momentum $J$, or has a critical spinning angular momentum $s_c$, a divergence of the center-of-mass energy $E_{cm}$ will be obtained.

gr-qc↗